Predicting the outcome of radiotherapy in brain metastasis by integrating the clinical and MRI‐based deep learning features

Abstract

Background A considerable proportion of metastatic brain tumors progress locally despite stereotactic radiation treatment, and it can take months before such local progression is evident on follow‐up imaging. Prediction of radiotherapy outcome in terms of tumor local failure is crucial for these patients and can facilitate treatment adjustments or allow for early salvage therapies. Purpose In this work, a novel deep learning architecture is introduced to predict the outcome of local control/failure in brain metastasis treated with stereotactic radiation therapy using treatment‐planning magnetic resonance imaging (MRI) and standard clinical attributes. Methods At the core of the proposed architecture is an InceptionResentV2 network to extract distinct features from each MRI slice for local outcome prediction. A recurrent or transformer network is integrated into the architecture to incorporate spatial dependencies between MRI slices into the predictive modeling. A visualization method based on prediction difference analysis is coupled with the deep learning model to illustrate how different regions of each lesion on MRI contribute to the model's prediction. The model was trained and optimized using the data acquired from 99 patients (116 lesions) and evaluated on an independent test set of 25 patients (40 lesions). Results The results demonstrate the promising potential of the MRI deep learning features for outcome prediction, outperforming standard clinical variables. The prediction model with only clinical variables demonstrated an area under the receiver operating characteristic curve (AUC) of 0.68. The MRI deep learning models resulted in AUCs in the range of 0.72 to 0.83 depending on the mechanism to integrate information from MRI slices of each lesion. The best prediction performance (AUC = 0.86) was associated with the model that combined the MRI deep learning features with clinical variables and incorporated the inter‐slice dependencies using a long short‐term memory recurrent network. The visualization results highlighted the importance of tumor/lesion margins in local outcome prediction for brain metastasis. Conclusions The promising results of this study show the possibility of early prediction of radiotherapy outcome for brain metastasis via deep learning of MRI and clinical attributes at pre‐treatment and encourage future studies on larger groups of patients treated with other radiotherapy modalities.

Analysis of Key Clinical Variables and Radiological Manifestations Associated with the Treatment Response of Patients with Brain Metastases to Stereotactic Radiosurgery

“…There are many complex factors affecting the treatment efficacy for BM [ 21 , 22 ], such as the basic situation of patients, the histological type and gene mutation of primary tumors, the control of primary tumors, the patient’s tolerance to chemotherapy, radiotherapy and targeted drugs, the degree of tumor response, the location, number and size of BM, the initial treatment method, and the presence of extracranial metastases, etc. It is quite difficult to cover all the influencing factors in one study.…”

Potential roles of transformers in brain tumor diagnosis and treatment

“…Deep AI learning is a promising data‐driven approach that can overcome this challenge [96] . Increased success using this approach has been reported for medical assessments, for instance, the characterization of brain metastases and the prediction of their outcomes [40] . Deep AI learning has recently achieved immense success in identifying primary and metastasized tumors and categorizing them according to their origin site based on whole‐slide histological data [97] .…”

“…A study by Jalalifar et al. revealed a novel DL framework to predict the local failure of SRT for treating brain metastases by utilizing data from the treatment‐planning MRI and pre‐treatment clinical details [40] . Their framework comprised a CNN (InceptionResNetV2) to identify textural features from 2D slices in the input MRI volumes and an long and short‐term memory (LSTM) network for the spatial dependency between the 2D slices.…”

A Self-Attention-Guided 3D Deep Residual Network With Big Transfer to Predict Local Failure in Brain Metastasis After Radiotherapy Using Multi-Channel MRI

“…Accordingly, the CNNs can potentially outperform the traditional radiomic models in diagnostic and prognostic applications for precision oncology by detecting patterns in medical images that are not captured by closed-form mathematical definitions of handcrafted radiomic features [28]- [30]. A recent publication from our group shows that the deep-learning features derived from 2D MRI slices outperform the standard clinical variables in predicting radiotherapy outcome in BM [31].…”